Developing an advanced, predictive battery health monitoring solution

In today’s world of advanced communication, industrial process control and automotive systems, the reliance on batteries to ensure power delivery is always available is becoming increasingly critical. Because of this, the ability to monitor the battery life-span and predict failure points is also becoming ever more important, but this is completely addressable with off the shelf microprocessor technology and software. In this paper, we will explore two different types of failure and lifespan prediction in battery systems: 1) Electrochemical Impedance Spectroscopy (EIS) and 2) Resistive measurements. Although both technologies can be implemented on their own to measure battery health, combining the two in a simple fashion on a low-cost microcontroller platform further increases the usefulness of the information collected. With many applications – automotive start-stop systems, advanced uninterruptable power supplies (UPS) for server and telecom applications and even hybrid automotive designs – predicting the failure point, available capacity or lifespan of battery back-up and power delivery systems is becoming more challenging for power electronics system designers. Especially in the case of UPS and telecom systems, the cost associated with the deliberate disconnection from AC mains, in conjunction with the long recharge cycle times that some of these batteries can have, drives the need for a more non-invasive fault prediction and monitoring solution. By taking advantage of the higher processing performance, system integration and communications peripherals of modern microcontrollers, one can develop a low cost, robust solution to fulfill this need based on EIS. This method, while completely non-invasive to the battery, offers a distinctly different perspective on the health and capacity of a battery system above and beyond simple current and voltage measurements. Although it is said that neither can match the absolute testing results of pure load testing, both are proving to be a low cost, reliable and time-effective solution or replacement. Brett Novak Marketing Manager C2000 microcontrollers Texas Instruments W H I T E P A P E R